Tulsa Law Review

Volume 48 | Issue 2 Article 17

2012

Constitutional Systems Theory: A ResearchAgenda Motivated by Vermeule, The System of theConstitution, and Epstein, Design for LibertyJenna Bednar

Follow this and additional works at: http://digitalcommons.law.utulsa.edu/tlrPart of the Law Commons

This Book Review is brought to you for free and open access by TU Law Digital Commons. It has been accepted for inclusion in Tulsa Law Review byan authorized administrator of TU Law Digital Commons. For more information, please contact daniel-bell@utulsa.edu.

Recommended CitationJenna Bednar, Constitutional Systems Theory: A Research Agenda Motivated by Vermeule, The System of the Constitution, and Epstein,Design for Liberty, 48 Tulsa L. Rev. 325 (2012).

Available at: http://digitalcommons.law.utulsa.edu/tlr/vol48/iss2/17

CONSTITUTIONAL SYSTEMS THEORY:A RESEARCH AGENDA MOTIVATED BY VERMEULE,THE SYSTEM OF THE CONSTITUTION AND EPSTEIN,

DESIGN FOR LIBERTY

Jenna Bednar*

ADRIAN VERMEULE, THE SYSTEM OF THE CONSTITUTION (2011). Pp. 220.Hardcover $35.00.

RICHARD A. EPSTEIN, DESIGN FOR LIBERTY: PRIVATE PROPERTY, PUBLICADMINISTRATION, AND THE RULE OF LAW (2011). Pp. 233. Hardcover $17.79.

ABSTRACT

This review considers Adrian Vermeule's The System of the Constitution1 and

Richard Epstein's Design for Liberty.2 Vermeule's contribution to the study of

constitutions applies some essential tenets of systems thinking that are barely evident in

legal analysis and almost nonexistent in political science, and so if the book is widely

read in both fields it has the potential to alter the trajectory of constitutional analysis for

the better. As (nearly) the opening move, the field remains wide open for future research,

and in this essay I offer some suggestions for new avenues. Epstein's concerns for the

fate of individual liberty and an expanding bureaucracy are evaluated in terms of

Vermeule's approach. Ultimately, what may rescue the United States from the over-

regulatory fate implied by Epstein's analysis is something not yet developed in

Vermeule's thesis: a constitutional culture, emerging within the public and defended by

it, which draws the line to preserve liberty and innovation.

I. INSTITUTIONAL ANALYSIS AND CONSTITUTIONAL SYSTEMS

Constitutions are important because they shape what is possible for a society,

whether it be economic growth, peace with its neighbors, or civic harmony. Or so we

think. A key scholarly interest in constitutional design is to develop an understanding of

how (and whether) the constitution affects the welfare of its subjects. Without question,

* Associate Professor of Political Science at the University of Michigan and author of THE ROBUSTFEDERATION: PRINCIPLES OF DESIGN (2009).

1. ADRIAN VERMEULE, THE SYSTEM OF THE CONSTITUTION (2011).2. RICHARD A. EPSTEIN, DESIGN FOR LIBERTY: PRIVATE PROPERTY, PUBLIC ADMINISTRATION, AND THE

RULE OF LAW (2011).

325

1

Bednar: Constitutional Systems Theory: A Research Agenda Motivated by Ver

Published by TU Law Digital Commons, 2012

TULSA LAW REVIEW

constitutions structure governments, and government performance influences - forbetter or worse - economic stability and growth, peace between nations (domestic orforeign), and liberty and social welfare. The problem with this theory is that theempirical correlation between constitutions and social outcomes is not neat; everyconstitutional form - democratic or authoritarian, majoritarian or consociational,presidential or parliamentary - has some examples of solid performance andsomewhere society suffers. To understand how and why constitutions are effective, wemust dig deeper into the constitutional form to consider how the components of theconstitution interact to shape social welfare.

In this essay I will review two recent books that examine the operation of theconstitution on social outcomes. For Vermeule, the contribution is in thinking about howthe components of government intersect to create a system, a constitutional order;3 inEpstein's case, he also presents a theory of the way that the institutions of governmentinteract, but does so to judge the product of the system. 4 Vermeule is positive, Epstein isnormative, and both offer interesting insights into the way that government affects socialwelfare. I will consider Vermeule first.

To understand Vermeule's contribution, it is necessary first to understand theessence of positive political theory ("PPT") because his book is rooted in it, butVermeule's main contribution is the way that he leaves PPT behind.5 PPT's intellectualancestry is in microeconomic theory.6 Like microeconomics, it begins with the analysisof individual agent behavior, and then from this foundation it builds a theory of socialoutcomes.7 The choices that individuals make are purpose-driven and (in mostinteresting cases) interdependent: what one agent chooses to do depends on what he orshe expects other agents to do. These decisions are often made with limited information,either about the preferences of other agents or about the consequences of one's actions.9

Social welfare is a product of the aggregation of individual decisions according to somerule of aggregation, and individual satisfaction with the social outcome is based on a ruleof distribution, or how the social good is shared among the members of the society. 10

For example, the U.S. presidential election is a product of individual actions -

votes - but different ways of compiling those votes can lead to different outcomes. TheU.S. Electoral College distorts the vote count in presidential elections, aggregating votesat the state level rather than nationally. Different electoral systems count votes(individual actions) in different ways. In turn, another function determines the value of asocial product to each individual. Budget surpluses or shortfalls are easily understood tohave differential distributional consequences: expenditures or revenue responsibilitiescan be targeted to hit different populations in different ways.

3. VERMEULE, supra note 1, at 3.4. See EPSTEIN, supra note 2, at 8.5. See VERMEULE, supra note 1, at 6-7.6. For an overview of microeconomic theory, see ANDREU MAS-COLELL, MICHAEL D. WHINSTON &

JERRY R. GREEN, MICROECONoMIc THEORY (1995).

7. See id at 1-16 (discussing the study of individual decision-making and the resulting assumptions fromthese decisions).

8. See id. at 11-14.9. See id at 170-71.

10. See id. at 117-21.

326 Vol. 48:325

2

Tulsa Law Review, Vol. 48 [2012], Iss. 2, Art. 17

http://digitalcommons.law.utulsa.edu/tlr/vol48/iss2/17

327CONSTITUTIONAL SYSTEMS THEORY

In PPT, the study of these rules of aggregation and distribution is referred to as

institutional analysis: these rules are institutions that determine the consequences of an

agent's action.I1 Therefore, we think of institutions as creating or shaping the incentive

environment. Different institutions will lead to different individual behaviors and,

therefore, to different collective outcomes.

All laws and all policies are institutions. Putting some bones on all of this

abstraction, institutions can be as specific as a law - say, the Equal Employment

Opportunity Act of 197212 ("EEOA") - or as broad as Congress. To simplify the

concepts of rules of aggregation and distribution, one might think of institutions as

encouraging behavior through carrots and sticks: inducements to encourage behavior or

punishments for failing to conform to the desired behavior. In the case of the EEOA, the

institution, a statute, provides incentives (and disincentives) to alter behavior; the

institution has a purpose, "to further promote equal employment opportunities for

American workers,, 13 and to reach that social outcome the EEOA mandates that

employers advertise positions and adopt hiring practices that are free from bias against a

variety of population categories, including race.14 Violations are punishable through

fines and related remedies.15

A law is itself a product of an earlier collective choice, that of individual

representatives acting in the institutional context of Congress. Viewing Congress as the

institution, it is a rule of aggregation (votes will be compiled in two separate assemblies,

requiring a majority of each, or a supermajority should the executive vetol6). It is also a

rule of distribution: each product - legislation - has distributional consequences for

the congressional members in terms of whether each member's constituents or campaign

backers are pleased with the legislation and the legislator's vote on it.

Given these foundations, institutions are associated with the production of political

goods. In economics, products are physical, whether pins or iPods. The goal is

efficiency. Productive efficiency is defined as a plan for the economy that produces the

most goods without requiring the input of any additional resources. In politics, products

are often less tangible and resources (apart from money) are almost always intangible.

Products may be measurable: poverty alleviation, growth, trade, budget deficits; or not:

liberty, justice, satisfaction. Many are only imperfectly measurable: peace, racial or

gender equity, representative responsiveness, transparency. One of the most

quantitatively elusive categories is also our most important: democracy. Valiant attempts

to codify relative levels do not end the comparative controversy with the United States

maxing out on the democracy index and yet still subject to important criticisms for

democratic failings. 17

11. See VERMEULE, supra note 1, at 15-16.12. Equal Employment Opportunity Act of 1972, Pub. L. No. 92-261, 86 Stat. 103 (codified as amended at

42 U.S.C. §§ 2000e-200e-17 (2012)).13. An Act, EEOC.GOv, http://www.eeoc.gov/eeoc/history/35th/thelaw/eco_1972.html (last visited Jan. 18,

2013).14. 42 U.S.C. § 2000e-2(a)(1) (2012).15. 42 U.S.C. § 2000e-2(n)(3) (2012).16. U.S. CONST. art. I, § 7, cl. 2.

17. See, e.g., WILLIAM H. RIKER, FEDERALISM: ORIGIN, OPERATION, SIGNIFICANCE 62-63, 142-43 (1964)

(arguing that federalism protects racist policies). See also ROBERT A. DAHL, How DEMOCRATIC IS THE

2012

3

Bednar: Constitutional Systems Theory: A Research Agenda Motivated by Ver

Published by TU Law Digital Commons, 2012

TULSA LAWREVIEW

To understand the effect of particular components of a constitution on itsperformance - evaluated by its correlation with the fulfillment of some social goal -PPT analysts often proceed as good microeconomic game theorists would: in thebehavioral equilibrium of a particular institutional context, what is the effect of varyingone of the inputs - one of the constitutional components? North and Weingast's seminalarticle in the field argued that it was the invention of a sovereign Parliament and anindependent judiciary that made the king more effective because once he was limited hecould credibly commit to repay money he would borrow.18 Therefore, lenders werewilling to extend credit, paying for the expansionist policies that built the BritishEmpire. 19 Other important examples of the institutional analysis of constitutions includeShugart and Carey, comparing presidential and parliamentary systems;20 Cox, analyzingelectoral rules; 2 1 Diermeier and Myerson on the effect of bicameralism on legislativeorganization;22 and Weingast, highlighting the role that federalism plays in promotingeconomic growth through market preservation. 23

Empirical evidence for some of the theoretical propositions is thin, perhapsbecause scholars sometimes mistakenly equate institutional functionality withinstitutional forms.24 More recent work has responded by expanding the scope ofanalysis; rather than varying one institution at a time, scholars trace the interdependenceof institutions. Independent judiciaries rely on competitive political environments,25 theefficacy of federalism depends upon the underlying distribution of authority betweenfederal and state governments,26 the aggregation of constitutional components produceseither divergent or consensual politics, which in turn has real effects on policy,27 andwhere policy stagnates as the number of constitutionally-provided veto players

AMERICAN CONSTITUTION? (2001), and Alfred C. Stepan, Federalism and Democracy: Beyond the US.Model, 10 J. DEMOCRACY 19 (1999), which both criticize the representation of the states in the Americanelectoral and legislative institutions coming at the expense of popular representation, and LAWRENCE LESSIG,REPUBLIC LOST: How MONEY CORRUPTS CONGRESS- AND A PLAN TO STOP IT (2011) (warning that specialinterest money biases democratic outcomes toward the privileged).

18. Douglass C. North & Barry R. Weingast, Constitutions and Commitment: The Evolution of InstitutionsGoverning Public Choice in Seventeenth Century England, 49 J. ECON. HIST. 803, 804-05 (1989).

19. Id. at 805.20. MATTHEW SOBERG SHUGART & JOHN M. CAREY, PRESIDENTS AND ASSEMBLIES: CONSTITUTIONAL

DESIGN AND ELECTORAL DYNAMICS 55-75 (1992).21. GARY W. Cox, MAKING VOTES COUNT: STRATEGIC COORDINATION IN THE WORLD'S ELECTORAL

SYSTEMS, 37-139 (1997).22. Daniel Diermeier & Roger B. Myerson, Bicameralism and Its Implications for the Internal

Organization ofLegislatures, 89 AM. ECON. REV. 1182 (1999).23. Barry R. Weingast, The Economic Role of Political Institutions: Market-Preserving Federalism and

Economic Development, 11 J. L. ECON. & ORG. 1 (1995).24. See, e.g., TORSTEN PERSSON & GUIDO TABELLINI, THE ECONOMIC EFFECTS OF CONSTITUTIONS (2003)

(taking care to state precisely what effects they measure to find support for a variety of more detailed claimsabout constitutional effects); Robert J. Barro, Democracy and Growth, I J. ECON. GROWTH 1 (1996) (findingmixed support for the theory that democracy promotes growth); Adam Przeworski & Fernando Limongi,Political Regimes and Economic Growth, 7 J. ECON. PERSP. 51 (1993) (same); Dani Rodrik, GettingInstitutions Right (Apr. 2004), available at http://www.wcfia.harvard.edu/sites/default/files/807 ifo-institutions%20article%20_April%202004_.pdf (arguing that effective institutional outcomes do not map intounique institutional designs).

25. Matthew C. Stephenson, "When the Devil Turns . : The Political Foundations for IndependentJudicial Review, 32 J. LEGAL STUD. 59, 61 (2003).

26. Weingast, supra note 23, at 4-5.27. See ROBERT D. COOTER, THE STRATEGIC CONSTITUTION 4-6 (2000).

328 Vol. 48:325

4

Tulsa Law Review, Vol. 48 [2012], Iss. 2, Art. 17

http://digitalcommons.law.utulsa.edu/tlr/vol48/iss2/17

CONSTITUTIONAL SYSTEMS THEORY

increases. 28

Perhaps the most interesting work, in terms of upside promise, is that which melds

the insights of interacting institutions with the legal culture, as Tushnet does in defining

the constitution as more than the text and the set of institutional arrangements to include

also the guiding principles for their operation.29 As these guiding principles change they

usher in new constitutional orders, as in the transition from a pursuit of justice through

law during the New Deal to Great Society period, to one that relies more heavily on

individual responsibility and market forces during the latter two decades of the twentieth

century.

II. VERMEULE, SYSTEM EFFECTS

It is against this backdrop of an emergent interest in institutional interdependence

that Vermeule steps forward with a series of works on constitutional system effects,

culminating in the book The System of the Constitution, under review here. 30

In a conceptual style reminiscent of Schelling3 1 or modem complex systems

theorists, Vermeule builds a vision of constitutionalism from the bottom up: individuals

interact with one another, their interactions shaped by a common institution. 32 In turn,

multiple institutions coexist, and their compiled force generates a constitutional order.33

Playing out within this vision of two-level systems comprising the constitutional order

are familiar concepts such as selection mechanisms and the influence of the invisible

hand,34 but also less familiar concepts, especially irreducibility and the theory of the

second best. 35 I will highlight these second two as I think that they have the most to offer

constitutional analysis.

A. Second-Best Institutionalism

With an emphasis on the second best and irreducibility, Vermeule makes a clear

break with his PPT compatriots. The theory of the second best, first described formally

by Lipsey and Lancaster,36 has been largely ignored by PPT scholars, even though it

may capture what they are groping toward, particularly when one invokes political

feasibility constraints. 37 The theory of the second best works as follows. Suppose there is

28. See GEORGE TSEBELIS, VETO PLAYERS: How POLITICAL INSTITUTIONS WORK 161-206 (2002).

29. MARK TUSHNET, THE NEW CONSTITUTIONAL ORDER 1 (2003) ("I think it productive to think about

constitutional orders in the United States that go beyond judicial doctrine and the written Constitution to

encompass relatively stable political arrangements and guiding principles.").

30. VERMEULE, supra note 1. Venneule is maddeningly productive. One can find him probing different

aspects of the general topic of constitutional system effects in a number of works over the past decade. Some of

my favorites include THE MECHANISMS OF DEMOCRACY: INSTITUTIONAL DESIGN WRIT SMALL (2007), LAW

AND THE LIMITS OF REASON (2009), and Forward: System Effects and the Constitution, 123 HARV. L. REv. 4

(2009).31. See THOMAS C. SCHELLING, MICROMOTIVES AND MACROBEHAVIOR (2d ed. 2006).

32. VERMEULE, supra note 1, at 3.33. Id.34. Id. at 65.35. See id. at 29-35.36. R. G. Lipsey & Kelvin Lancaster, The General Theory ofSecond Best, 24 REV. ECON. STUD. 11 (1956-

1957).37. Illuminating exceptions include Bruce Talbot Coram, Second Best Theories and the Implications for

Institutional Design, in THE THEORY OF INSTITUTIONAL DESIGN 90 (Robert E. Goodin ed., 1996); JON ELSTER,

3292012

5

Bednar: Constitutional Systems Theory: A Research Agenda Motivated by Ver

Published by TU Law Digital Commons, 2012

TULSA LAW REVIEW

some determination about the optimal mix of institutions for governing a population: say,in order to promote economic growth, one follows Weingast's prescriptions for market-preserving federalism.38 The institutional prescription includes a federal system wherestates set fiscal policy and the national government protects property rights. 39 Supposethat in order to maintain this balance, an independent judiciary is necessary to umpire theboundaries of federalism.40 Suppose Stephenson is right and an independent judiciaryrequires a competitive political system. 4 1 Summing it together, in our hypotheticalexample we would have an optimal institutional prescription of federalism, anindependent judiciary, and a competitive party system.

With a prescription for an optimal institutional arrangement designed for fiscalrestraint in hand, one can ask: What happens if the party system is not competitive, if oneparty dominates? That is, more generally, what happens to the components in ouroptimal system if one component is unavailable? The theory of the second best says thatit is possible that if one variable is constrained (here, the competitive party system) thenthe second best institutional arrangement may require alterations to the other variables. 42

In our hypothetical case, the lack of a competitive party system may make anindependent judiciary unobtainable, and if there is no functional replacement for thejudiciary, then federalism may need to be abandoned and the economy managedcentrally.

Vermeule invokes the theory of the second best to emphasize its corollary: theapproximation fallacy.43 If one cannot achieve the optimal arrangement of institutions,then one might think that the best strategy is to set the institutional arrangement to be asclose as possible to optimal. In our hypothetical, the constitutional designer would forgeahead with federalism and set up a court and hope for the best in the absence of thecompetitive party system. That approach would be theoretically misguided andpotentially disastrous in practice. If one component is constrained, or even missing, thenit could cause another component to behave in an unanticipated way. Returning to theexample above, if political competition is necessary for an independent judiciary, andthere is no political competition, then having an ineffective judiciary - or one beholdento the national government - may cause greater problems than if judicial safeguards offederalism were absent and federal skirmishes were more transparently the purview ofpolitics.44

Vermeule contrasts the systems theory approach to one that he dubs "piecemeal":criticisms of the American system as being undemocratic or unworkable that point to the

EXPLAINING SOCIAL BEHAVIOR: MORE NUTS AND BOLTS FOR THE SOCIAL SCIENCES (2007); Dani Rodrik,Second-Best Institutions, 98 AM. ECON. REV. 100 (2008); Adrian Vermeule, Hume's Second-BestConstitutionalism, 70 U. CHI. L. REV. 421 (2003).

38. See Weingast, supra note 23, at 3.39. Id. at 4-5.40. See id at 19-21.41. Stephenson, supra note 25, at 61.42. VERMEULE, supra note 1, at 29-30 ("[T]he failure to satisfy optimality conditions on one variable

means that other variables must take on suboptimal values as well, in order to compensate for the initialfailure.").

43. Id. at 30.44. For more examples along these lines, see my book THE ROBUST FEDERATION, supra note *, at 147-68.

330 Vol. 48:325

6

Tulsa Law Review, Vol. 48 [2012], Iss. 2, Art. 17

http://digitalcommons.law.utulsa.edu/tlr/vol48/iss2/17

CONSTITUTIONAL SYSTEMS THEORY

democratic deficit or inefficiencies in virtually every component of the system.4 5

Vermeule points out that if one diagnoses the ills of democracy in the United States

component-by-component, one is naturally inclined to attempt to improve the

functioning of democracy in whatever way possible.46 Incremental improvements

include increasing the number of elected offices, reducing the number of elections,

adding regulation that encourages party organizations (or not), or expanding judicial

elections. Each of these adjustments may have effects that spill over to other system

components.The piecemeal approach implicitly ascribes to a linear model, where the sum of the

parts is exactly equal to the whole. In a linear model, marginal improvement to one

component necessarily improves the whole. In a complex system, where the components

are interdependent, aggregation effects can be non-linear; incremental improvements

could make the whole much better than the marginal change to the component. But it

could also make it worse, perhaps much worse.47 While the problem is appreciated in

economics, it is perhaps especially an issue in political or legal systems because of the

non-independence of the institutions.Institutional arrangements may be twitchy: political and legal systems may be

particularly prone to discontinuities that would create large leaps in the institutional

space between first and second best institutional arrangements. This observation leads to

two implications worthy of future study. First, constitutions should not be written by a

subcommittee, with one group writing the section on the judiciary and another on the

legislature, or even one writing on the institutions of governance and another on

individual rights. As a system, the constitution must be appreciated as a whole.

Second, if one institutional structure is suddenly rendered ineffective, it can have

massive reverberations throughout the system. For this reason, constitutional scholars

need to study resilience of constitutional systems. Resilience studies should include

knock-out experiments: what happens to the rest of the system if one component is

rendered ineffective? Knock-out experiments are common to ecological studies and

uncommon, perhaps to the point of absent, in constitutional studies.

B. Irreducibility

Systems theorists accept that the system may have properties that are distinct from

its components. Vermeule again phrases this lesson in terms of a warning: one must not

be tempted to infer that because a system has particular characteristics - say,

transparency, or democratic responsiveness - that each component of the system will

have that characteristic. 48 It is more intuitive to accept that the whole may differ from the

parts, which perhaps is why Vermeule does not need to spend much space illustrating

that phenomenon. The canonical example of irreducibility is the existence of the

unelected judiciary within the American democracy. It is the sort of contradiction that

confounds the public and great scholars alike: an unelected, unrepresentative,

45. VERMEULE, supra note 1, at 6-7.46. Id. at 55.47. See id.48. Id. at 27.

3312012

7

Bednar: Constitutional Systems Theory: A Research Agenda Motivated by Ver

Published by TU Law Digital Commons, 2012

TULSA LAW REVIEW

unaccountable body of nine can block the intentions of the public's political agents. Andyet, the judiciary is heralded as one of our democratic system's three key pillars, with apopularity rating regularly triple that of Congress. 4 9

An interesting offshoot from the concept of irreducibility is that it affects how wethink not only about individual components, but about individual agents within anyinstitution. Consider judges: a sensitivity to irreducibility ought to shape our evaluationof individual judges. Just as no individual can be diverse - it does not make sense to sayit - Vermeule would argue on the same grounds that no individual can be intrinsicallyvaluable. 50 The value of any public official (here, judges, particularly) depends on theothers in the group. 5 1 A determination of the desirability and efficacy of any one judgerequires consideration of the full group.52 Recruitment and selection strategies cannot bebased upon individual characteristics alone, but how those characteristics fit within thefuller system, whether it be a set of individuals or institutions. The beauty of a star isbased on its position relative to others within a constellation. Systems emphasizerelational context.

Selection effects are a terrific trope that Vermeule uses to illustrateinterdependence.53 Vermeule also develops an interesting thesis regarding the way thatconstitutional rules generate a selection mechanism that can change the set of candidatesfor public positions.54 He introduces the theory of a screening mechanism, which acts asa filter on the whole pool of potential applicants (or candidates). 55 Like the irreducibilityand second order effects, the theory is most interesting when it generates nonobviousconclusions. He focuses particularly on the effect of compensation rules.56 TheConstitution has very similar rules about compensation for Congress and the judiciary,with one key difference: in Congress, sitting members cannot vote themselves a raise. 5 7

They can vote for a raise for future Congresses.58 This mechanism clearly disincentivizesvoting for raises, which of course was its purpose. 59 But as long as judicial raises are tiedto congressional ones, the amendment exacerbates the unintentional consequences of thecompensation rules.60 Judicial salaries are falling even farther behind the compensation

in private practice.61 The theory of selection mechanisms complements the theories ofirreducibility and second order effects well.

49. Adam Liptak, & Allison Kopicki, Approval Ratings for Justices Hits Just 44% in New Poll, N.Y.TIMES, June 8, 2012, at Al, available at http://www.nytimes.com/2012/06/08/us/politics/44-percent-of-americans-approve-of-supreme-court-in-new-poll.html?pagewanted=all& r-0 (noting that even though theSupreme Court's current approval rating is low, it is still higher than Congress's approval rating).

50. See VERMEULE, supra note 1, at 111-12.51. Id at 110-11.52. See id. at 132.53. See id at 131-33.54. Id. at 123-31.55. Id. at 114-16,123.56. Id at 124-25.57. See U.S. CONST. art. III, § 1; VERMEULE, supra note 1, at 124.58. VERMEULE, supra note 1, at 124.59. Id.60. Id.61. See id.

332 Vol. 48:325

8

Tulsa Law Review, Vol. 48 [2012], Iss. 2, Art. 17

http://digitalcommons.law.utulsa.edu/tlr/vol48/iss2/17

CONSTITUTIONAL SYSTEMS THEORY

C. Further applications

I would like to highlight two more applications. Vermeule skewers hopes of

accountability through the ballot box, but then revives hopes by introducing an

alternative: a self-regulating system of internal accountability.62 In the complexity of the

American system, one might already hold the opinion that electoral accountability -

causing public officials to serve the public interest or be removed from office - is

inefficient at best, and perhaps nonexistent, given that the vote itself is a very crude

message, the lackadaisical attitude of most Americans to all but the most salient electoral

choices, and the ability of special interests to control information. Vermeule squashes

any lingering glimmer of hope that electoral accountability is possible.63 The

components are too interrelated to assign responsibility accurately. 64

In fact, Vermeule goes even farther: not only is it hopeless to imagine that electoral

accountability is possible, but it is explicitly not so by design. 65 In a second-order

system, no political agent can be evaluated independent of the other agents.66 Taken to

its extreme form, the Vermeule thesis even might refuse to accept graft, abuse of power,

or golfing through the workday as an indication that an individual ought not be reelected:

perhaps the graft causes him to work harder or realigns the interests of others

beneficially; perhaps power-grabbing improves the orientation of authority; perhaps

society is better off when a political agent is inactive, at least during a Wednesday

afternoon golf league. Vermeule offers another possibility for accountability: through

internal regulation or self-regulation. 67 The components may effectively check one

another; Madison's "auxiliary precautions" may in fact be what maintains our

democracy.68 Or, it may be that popular safeguards are inactive for ordinary politics;with a theory of institutional complementarity (another element in systems analysis),they may be inactive but the possibility of electoral punishment makes other safeguards

- judicial, political, and the fragmented structure of government - more effective then

they would be absent effective electoral constraints. 69

Vermeule provides us a framework for judging the efficacy of one system against

another.70 Dahl and Stepan reject the complexity of the American system, ultimately dueto one principle: it fails on electoral accountability grounds.7 1 Rather than emphasizing

the failure of each component - given Vermeule's convincing work that is insufficient

criterion for rejecting the whole system - instead, their objection should be phrased in

terms of the agency of the individual; in a democratic society, the individual should feel

authority over government. In which system, the British, unified and majoritarian, or the

American, fragmented and federal, does policy more closely align with the public will?

62. See id. at 123-31.63. Id. at 131-33.64. See id at 133.65. Id. at 132.66. Id. at70.67. Id. at 117-23.68. See id. at 38-43.69. See THE ROBUST FEDERATION, supra note *, at 147-212.70. See VERMEULE, supra note 1, at 14-37.71. DAHL, supra note 17, at 99-119; Stepan, supra note 17.

3332012

9

Bednar: Constitutional Systems Theory: A Research Agenda Motivated by Ver

Published by TU Law Digital Commons, 2012

TULSA LAW REVIEW

The instruments that drive the alignment differ between the two cases: on the one hand,there is electoral accountability, supported through responsible parties; on the other, thesystem is self-regulating. This review is not the place to conduct this comparison, but theproper way to proceed, it seems to me, is to consider both the capacity and likelihood ofboth at heeding the public interest, but also, and quite importantly, the likelihood thateither would fail to reach it and perhaps fail egregiously. That is, the appropriate methodof comparison would consider the resilience of the system; its ability to resist tyranny orexploitation for private gain. I suspect that a full analysis will conclude that there is noabsolute ranking of the two systems - that their performance is conditional, and itwould be most helpful to know those conditions.

Perhaps Vermeule's most ambitious application is a fresh interpretation of thechanges in the distribution of federal and state authority from the nineteenth to thetwentieth centuries.72 He starts with a few cases that, if developed fully, could beinteresting alternatives within American political development scholarship.73 Theinterpretation rests on a theory of self-stabilizing and self-undermining selection rules.Vermeule links the expansion of federal power, noted by many students of federalism,not as predetermined constitutionally (as Riker argues 74), not as fatalistically inevitable,and not as a product of the federal government intentionally stealing power from thestates. Instead, he seeks an explanation rooted in the incentives of the individual,augmented by selection effects. 75 Industrialization, war, and the Depression created anational perception that the federal government was more effective than the states atsolving the nation's great problems.76 The opportunity for enhanced power at the federallevel led candidates who were individually power hungry to seek office; they valued notonly the salary, but the power.77 My assessment: I do not know if I buy it, but it is aprovocative, original explanation, and it is extremely hard to find in a well-trodliterature.

Vermeule raises two topics that read as placeholders for future research. The firstis the theory of norms, including their emergence. In chapter three, on invisible hand,Vermeule addresses the theory of norms. 79 While parts of the chapter are excellent, like

80the distinction between designing an institution and designing an outcome, overall, theclarity of the other chapters is lacking here. The good news is that Vermeule maycontinue to work on it, because by invoking theorists like Tushnet, he makes clear thatthe effect, emergence, and evolution of norms is key to a full theory of constitutionalorder, where conventions ride equally alongside law. 8 1

72. See VERMEULE, supra note 1, at 161-68.73. See id. at 28-36.74. RIKER, supra note 17, at 11-25.75. VERMEULE, supra note 1, at 29-37.76. See id. at 38-43.77. See id. at 11-16.78. Id. at 66.79. Id80. See id at 79-80.81. See id. at 94- 100.

334 Vol. 48:325

10

Tulsa Law Review, Vol. 48 [2012], Iss. 2, Art. 17

http://digitalcommons.law.utulsa.edu/tlr/vol48/iss2/17

CONSTITUTIONAL SYSTEMS THEORY

A second topic ripe for expansion is the theory of constitutional evolution.82 If one

stitched pieces together from the book, one would have a decent foundation for a new

theory of constitutional change. To explain change, Vermeule switches gears from

emphasizing the distinctness between components and system at one point in time to

their coevolution.83 The system, after all, is built by the components, so it makes sense

that the system would change as components change, although that change may be

unanticipated and happen much more quickly than one would expect. To theorize fullyabout constitutional evolution using systems analysis, one will need to map out

theoretically and empirically how the components are connected - how loose or tight

they are - and develop a theory of feedback within constitutions to understand how the

constitutional system might tip from one state to another.84 It is quite possible that norms

or constitutional culture (Tushnet's principles 85) are what bind components together. Ifso, given that much constitutional change comes through reinterpretation, 86

understanding the diffusion of constitutional ideas will be necessary in building thetheory of constitutional change.

The book is first and foremost a work of constitutional theory, but to show the

application of these concepts, Vermeule peppers the text with quick examples developed

in a few sentences or sometimes extending to a few pages. 87 1 have highlighted a few inthis essay. None are meant to represent definitive empirical support of his model, butinstead are used to illustrate the logic through familiar examples. The sheer number ofillustrations invites the reader to think of many more. Vermeule tends to stick closely to

the relationship between political institutions and questions of representation, makingonly brief remarks about rights (freedom of the press, voting rights, especially of black

Americans). But one can imagine applications to gender equality, marriage equality,privacy, war powers, emergency powers, etc. It is the best kind of creative force, one thatbubbles over with enthusiasm for the work of others and tosses in so many ideas andexamples that the reader cannot help but get caught up in the act and generate ideas and

examples of her own.

III. EPSTEIN

An important feature of Vermeule's argument is the pains that he takes to

emphasize that theory is non-deterministic and that the American system, in particular,

may be dysfunctional.89 After a fairly bullish beginning, laying out why non-democraticcomponents can create a democratic system (and even may be necessary for one, in thecase of countermajoritarian judicial review90), Vermeule backs away from what might

82. See id. at 97-99.83. See id. at 102-33.84. See P. J. Lamberson & Scott E. Page, Tipping Points, 7 Q. J. POL. SC. 175 (2012) (outlining the

concept of constitutional tipping).85. TUSHNET, supra note 29, at 8-33.86. See Sanford Levinson, Designing an Amendment Process, in CONSTITUTIONAL CULTURE AND

DEMOCRATIC RULE 271 (John Ferejohn, Jack N. Rakove & Jonathan Riley eds., 2001).87. VERMEULE, supra note 1, at 22-23, 92-93, 115.88. Id. at90,118-19.89. See id. at 36-37, 63-64, 94-100, 131-33.90. See id. at 58-63.

2012 335

11

Bednar: Constitutional Systems Theory: A Research Agenda Motivated by Ver

Published by TU Law Digital Commons, 2012

TULSA LAW REVIEW

have been interpreted as a case for the optimality of the American system. 9 1 It was wiseto withdraw, for too many smart people like Epstein would shoot arrows his way.

Epstein makes the case that the American government has grown beyond itsusefulness, and such growth is not only undesirable but inevitable.92 While the argumentis cloaked in strong normative overtones - Epstein is a libertarian, and so perhaps onecould see the conclusion coming - it is premised upon a significant claim: theconstitutional system is incapable of self-limitation. 93 Regulation introduces discretion,and public administrations cannot manage discretion well. 94 Regulatory agencies willgrow, stifling the innovation that makes continued growth possible. 95

Limited government, in Epstein's model of the constitutional system, dependsupon effective judicial monitoring, and Epstein recognizes that judges are inevitablyimperfect, although he wishes it were not so.96 He writes: judges should take"exceptional care to guard against actual bias," 97 a sweet sentiment that can go exactlynowhere. Epstein's dependence on the elusive rule of law echoes Hayek, where anindependent judiciary is a citizen's last line of defense against an encroachingbureaucracy.98 In essence, Epstein makes the opposite argument of Wittman, who arguedthat economic markets are no more efficient than political markets, and that the set ofinstitutions in the American democracy protect liberties and property just as well asmarket could.9 9 Epstein flatly disagrees.1 00

There are several ways that we might use the Vermeule book to consider Epstein'sconjecture. Vermeule would counter first that given the rule of irreducibility, it ispossible to have a functional system even if the parts are (somewhat) broken. 101 Epsteinwould use Vermeule against himself here: the claim is only one of existence, notinevitability; just because a well-functioning system is possible does not mean that it willbe obtained (or, one should add, maintained).102

Vermeule would respond (along with a chorus of other legal theorists) that thecitizen is protected in many ways, not just through the courts.103 There are alternativesthat might counter a regulatory zeal, including the people themselves. The rise of the TeaParty is a recent example of popular safeguards at work.104 Perhaps it is the lack of

91. See id at 145-49.92. EPSTEIN, supra note 2, at 5-6.93. Id. at 6-7.94. Id.95. Id. at 7. There is, in Epstein's addition, the notable absence of the justification for regulation. When it

corrects market imbalances, regulation can be socially productive, a counterweight that might bring the netbenefit of regulation above zero, despite administrative costs and other inefficiencies.

96. Id. at 18.97. Id.98. EPSTEIN, supra note 2, at 17-18; F. A. HAYEK, THE CONSTITUTION OF LIBERTY 208-14 (Univ. of Chic.

Press 1960).99. DONALD A. WITTMAN, THE MYTH OF DEMOCRATIC FAILURE: WHY POLITICAL INSTITUTIONS ARE

EFFICIENT 87-114 (1995).100. See EPSTEIN, supra note 2, at 55-57.101. See VERMEULE, supra note 1, at 29-37.102. See EPSTEIN, supra note 2, at 16-17.103. See VERMEULE, supra note 1, at 117-31.104. For a discussion of the rise of the Tea Party, see Rebecca E. Zietlow, Popular Originalism? The Tea

Party Movement and Constitutional Theory, 62 FLA. L. REv. 483, 488-91 (2012).

336 Vol. 48:325

12

Tulsa Law Review, Vol. 48 [2012], Iss. 2, Art. 17

http://digitalcommons.law.utulsa.edu/tlr/vol48/iss2/17

CONSTITUTIONAL SYSTEMS THEORY

reliability of the political system - and the recent strategic fizzling of the Tea Party,

taking the Republican Party down with it105 - that causes assurances of the sufficiency

of popular and political safeguards to fall flat.

Vermeule might also suggest (again, continuing to accept the claim that the United

States is overregulated), that we are in a bad cycle, using the selection mechanism part of

Vermeule's theory. Vermeule focuses on compensation, qualifications, and on incentives

at the individual level.106 The application requires deeper analysis, but it is possible that

feedback in the system leads to overregulation. The strongest application relating

individual incentives to regulation points in the opposite direction: the revolving door

theory of bureaucracy says that bureaucrats move in and out of industry and anti-

regulatory lobbying agencies. Career incentives dampen the bureaucrat's enthusiasm for

regulation. However, to support Epstein, they may channel regulatory efforts in a biased

way, one that is socially inefficient.Given Vermeule's treatment of constitutions as systems, Epstein offers a useful

counterpoint. I think it serves as a reminder of the importance in understanding not just

the institutional components of a constitutional order, but the guiding principles as well.

Ultimately, a full appreciation of the American system, for better or for worse, will only

come once we have a proper theory of constitutional resilience. With it, we might either

confirm or reject Epstein's conjecture.

IV. FINAL THOUGHTS

Science is cumulative, and so good work does not aim to answer all questions, but

instead, opens new pathways for research. Work should be judged at least as much based

upon what is now possible because this work exists as upon what the work answers. I

titled this review a Proposed Research Agenda, which I mean. It may be easy to see

where I see intellectual potential to move forward with a systems theory approach to the

study of constitutions given Vermeule's opening waltz and the opportunities for applying

systems thinking to evaluate claims of constitutional failure, as Epstein offers. Systems

thinking brings tools and a perspective that encourages scholars to note the interactive

effects; sometimes interactions make institutions even more powerful, sometimes less so,

sometimes just different. Taking full measure of the theory of second best institutions

ought to be the standard practice in PPT. Going beyond the immediate text of these

books, interesting questions concern institutional codependence and complementary and,

perhaps most tantalizing, a fresh understanding of legal evolution and dynamics, which

can lead us toward an understanding of constitutional resilience.

105. See Ariel Edwards-Levy, Tea Party Hits Record Level of Unpopularity in Public Opinion Poll, THE

HUFFINGTON POST (Jan. 7, 2013, 2:34 PM), http://www.huffingtonpost.com/2013/01/07/tea-party-poll n_2425833.html.

106. See VERMEULE, supra note 1, at 101-17.

3372012

13

Bednar: Constitutional Systems Theory: A Research Agenda Motivated by Ver

Published by TU Law Digital Commons, 2012

&

14

Tulsa Law Review, Vol. 48 [2012], Iss. 2, Art. 17

http://digitalcommons.law.utulsa.edu/tlr/vol48/iss2/17